© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 2-RES-JB
    6. 823
    Prev
    Next
    INDEX
    Explanations

    the adjective 'malicious'

    oai_token-act-pair · gpt-4-turbo

    instances of malicious actions or intentions

    oai_token-act-pair · gpt-3.5-turbo

    mentions of harmful or malicious intent

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ĸļ
    -1.10
    arist
    -0.85
    hetti
    -0.82
    ILA
    -0.70
    ļéĨĴ
    -0.69
    ills
    -0.69
    quart
    -0.68
    oran
    -0.67
     Gap
    -0.67
     Browne
    -0.67
    POSITIVE LOGITS
     mischief
    0.94
     malicious
    0.94
     intent
    0.90
    ly
    0.90
    icious
    0.88
    vertising
    0.86
     behavi
    0.81
     behaviour
    0.80
     payload
    0.80
     guiActiveUn
    0.75
    Activations Density 0.005%

    No Known Activations